linux-kernel - [PATCH v3 06/10] perf top: Switch user option to use BPF filter

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250425214008.176100-7-irogers@google.com>
Date: Fri, 25 Apr 2025 14:40:04 -0700
From: Ian Rogers <irogers@...gle.com>
To: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>, 
	Arnaldo Carvalho de Melo <acme@...nel.org>, Namhyung Kim <namhyung@...nel.org>, 
	Mark Rutland <mark.rutland@....com>, 
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>, Jiri Olsa <jolsa@...nel.org>, 
	Ian Rogers <irogers@...gle.com>, Adrian Hunter <adrian.hunter@...el.com>, 
	Kan Liang <kan.liang@...ux.intel.com>, James Clark <james.clark@...aro.org>, 
	Dapeng Mi <dapeng1.mi@...ux.intel.com>, Thomas Richter <tmricht@...ux.ibm.com>, 
	Veronika Molnarova <vmolnaro@...hat.com>, Hao Ge <gehao@...inos.cn>, 
	Howard Chu <howardchu95@...il.com>, Weilin Wang <weilin.wang@...el.com>, 
	Levi Yun <yeoreum.yun@....com>, "Dr. David Alan Gilbert" <linux@...blig.org>, 
	Dominique Martinet <asmadeus@...ewreck.org>, Xu Yang <xu.yang_2@....com>, 
	Tengda Wu <wutengda@...weicloud.com>, linux-perf-users@...r.kernel.org, 
	linux-kernel@...r.kernel.org, bpf@...r.kernel.org
Subject: [PATCH v3 06/10] perf top: Switch user option to use BPF filter

The existing --uid option finds user processes by scanning
/proc. Scanning /proc is inherently racy due to processes being in
/proc but then exiting before perf does the perf_event_open. When the
perf_event_open fails perf will terminate reporting errors which is a
disappointing user experience. Scanning /proc when perf starts also
cannot inform perf of new user processes starting.

The ability to filter perf events with BPF isn't new, and has been in
the perf tool for 10 years:
https://lore.kernel.org/all/1444826502-49291-8-git-send-email-wangnan0@huawei.com/
An ability to do filtering on the command line with a BPF program
that's part of perf was added 2 years ago:
https://lore.kernel.org/all/20230314234237.3008956-1-namhyung@kernel.org/
This was then extended to support uids as a way of filtering:
https://lore.kernel.org/all/20240524205227.244375-1-irogers@google.com/

This change switches the perf top --uid option to use the BPF filter
code to avoid the inherent race and existing failures.

Using BPF has permission issues in loading the BPF program not present
in scanning /proc. As the scanning approach would miss new programs
and fail due to the race, this is considered preferable. The change
also avoids opening a perf event per PID, which is less overhead in
the kernel.

Signed-off-by: Ian Rogers <irogers@...gle.com>
---
 tools/perf/builtin-top.c | 22 ++++++++++++----------
 tools/perf/util/top.c    |  4 ++--
 tools/perf/util/top.h    |  1 +
 3 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index f9f31391bddb..8890bec9b63c 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -642,7 +642,7 @@ static void *display_thread_tui(void *arg)
 	 */
 	evlist__for_each_entry(top->evlist, pos) {
 		struct hists *hists = evsel__hists(pos);
-		hists->uid_filter_str = top->record_opts.target.uid_str;
+		hists->uid_filter_str = top->uid_str;
 	}
 
 	ret = evlist__tui_browse_hists(top->evlist, help, &hbt, top->min_percent,
@@ -1566,7 +1566,7 @@ int cmd_top(int argc, const char **argv)
 		    "Add prefix to source file path names in programs (with --prefix-strip)"),
 	OPT_STRING(0, "prefix-strip", &annotate_opts.prefix_strip, "N",
 		    "Strip first N entries of source file path name in programs (with --prefix)"),
-	OPT_STRING('u', "uid", &target->uid_str, "user", "user to profile"),
+	OPT_STRING('u', "uid", &top.uid_str, "user", "user to profile"),
 	OPT_CALLBACK(0, "percent-limit", &top, "percent",
 		     "Don't show entries under that percent", parse_percent_limit),
 	OPT_CALLBACK(0, "percentage", NULL, "relative|absolute",
@@ -1757,15 +1757,17 @@ int cmd_top(int argc, const char **argv)
 		ui__warning("%s\n", errbuf);
 	}
 
-	status = target__parse_uid(target);
-	if (status) {
-		int saved_errno = errno;
-
-		target__strerror(target, status, errbuf, BUFSIZ);
-		ui__error("%s\n", errbuf);
+	if (top.uid_str) {
+		uid_t uid = parse_uid(top.uid_str);
 
-		status = -saved_errno;
-		goto out_delete_evlist;
+		if (uid == UINT_MAX) {
+			ui__error("Invalid User: %s", top.uid_str);
+			status = -EINVAL;
+			goto out_delete_evlist;
+		}
+		status = parse_uid_filter(top.evlist, uid);
+		if (status)
+			goto out_delete_evlist;
 	}
 
 	if (target__none(target))
diff --git a/tools/perf/util/top.c b/tools/perf/util/top.c
index 4db3d1bd686c..b06e10a116bb 100644
--- a/tools/perf/util/top.c
+++ b/tools/perf/util/top.c
@@ -88,9 +88,9 @@ size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size)
 	else if (target->tid)
 		ret += SNPRINTF(bf + ret, size - ret, " (target_tid: %s",
 				target->tid);
-	else if (target->uid_str != NULL)
+	else if (top->uid_str != NULL)
 		ret += SNPRINTF(bf + ret, size - ret, " (uid: %s",
-				target->uid_str);
+				top->uid_str);
 	else
 		ret += SNPRINTF(bf + ret, size - ret, " (all");
 
diff --git a/tools/perf/util/top.h b/tools/perf/util/top.h
index 4c5588dbb131..04ff926846be 100644
--- a/tools/perf/util/top.h
+++ b/tools/perf/util/top.h
@@ -48,6 +48,7 @@ struct perf_top {
 	const char	   *sym_filter;
 	float		   min_percent;
 	unsigned int	   nr_threads_synthesize;
+	const char	   *uid_str;
 
 	struct {
 		struct ordered_events	*in;
-- 
2.49.0.850.g28803427d3-goog